[comp.arch] Bitblt chips and things

erict@flatline.UUCP (j eric townsend) (07/15/88)

In article <1090@ficc.UUCP>, peter@ficc.UUCP (Peter da Silva) writes:
> In article <76700032@p.cs.uiuc.edu>, gillies@p.cs.uiuc.edu writes:
> > 1.  Simple programs that need OPTIMAL speed, and would be outrageously
> > expensive on ANY machine.  The major example is BITBLT, a subroutine
> > with about 10-15 parameters that does a massive amount of linear-time
> > work.  If the BITBLT subroutines generates the machine code expressly

> You can save an even greater factor by building the BitBlt into the
> hardware. This is what the Commodore Amiga does, using a chip called
> the Biltter. I'm pretty surprised that a chipset this good that's cheap


Isn't this in essence what CBM did with the "sprite"s on the Commodore
64?  (Can I mention that machine in this group?  I didn't think so. :-)

For those that never toyed with the 64, it has a graphics chip that
handles from 0-8(? I think 8) MOB's called "sprites" (stupid name).
You tell it what the MOB looks like, then just tell it to move the
MOB.  It handles multi-color (out of 16), foreground/background, and
automatic collision detection.  I was amazed to find that IBM PC's,
Apple II's, and the other early C64 competition didn't have this capability.

Anybody know what home/micro computer did it first?  I'm pretty
sure the Vic-20 could do this, but I'm not sure about the PET.

-- 
Skate UNIX or go home, boogie boy...
     [Obscure joke goes here]
J. Eric Townsend ->uunet!nuchat!flatline!erict smail:511Parker#2,Hstn,Tx,77007
             ..!bellcore!tness1!/

lbruck@eneevax.UUCP (Lewis S.Bruck) (07/15/88)

In article <1110@flatline.UUCP> erict@flatline.UUCP (j eric townsend) writes:
>In article <1090@ficc.UUCP>, peter@ficc.UUCP (Peter da Silva) writes:
>> hardware. This is what the Commodore Amiga does, using a chip called
>> the Biltter. I'm pretty surprised that a chipset this good that's cheap
>
>Isn't this in essence what CBM did with the "sprite"s on the Commodore
>64?  (Can I mention that machine in this group?  I didn't think so. :-)
>
>For those that never toyed with the 64, it has a graphics chip that
>handles from 0-8(? I think 8) MOB's called "sprites" (stupid name).
>You tell it what the MOB looks like, then just tell it to move the
>MOB.  It handles multi-color (out of 16), foreground/background, and
>automatic collision detection.  I was amazed to find that IBM PC's,
>Apple II's, and the other early C64 competition didn't have this capability.
>
>Anybody know what home/micro computer did it first?  I'm pretty
>sure the Vic-20 could do this, but I'm not sure about the PET.
>

I believe tha the first (or at least one of the earliest) was the old 
ATARI 400/800, which had a system called player/missile graphics which had
quick movement independent of actually manipulating display memory and
collision detection.  The chipset was designed by Jay Miner, who also designed
the Amiga's chipset.

Sprite (or PM or MOB) graphics are different from the blitting that is done
on the Amiga.  The sprites are independent of the display memory, acting as
an overlay, while the blitter actually manipulates display memory (and any
other memory in it's address space).  While the sprites are faster, they are
limited in terms of colors and width of image, while the blitter is a more
general purpose device.

------------------------------------------------------------------
Lewis Bruck
lbruck@eneevax.umd.edu
Not officially representing anybody, anywhere, for any reason

chris@mimsy.UUCP (Chris Torek) (07/15/88)

In article <1110@flatline.UUCP> erict@flatline.UUCP (j eric townsend) writes:
>Isn't this in essence what CBM did with the "sprite"s on the Commodore
>64?  (Can I mention that machine in this group?  I didn't think so. :-)

But are we allowed to reminisce about intra-vertical-trace interrupts
and adjusting the sprites and colors on the fly to get more sprites
and colors than the hardware supported?  (Talk about your realtime
applications! :-) )

(No, actually, I never did any of that.)
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163)
Domain:	chris@mimsy.umd.edu	Path:	uunet!mimsy!chris

dlm@druhi.ATT.COM (Dan Moore) (07/18/88)

in article <1110@flatline.UUCP>, erict@flatline.UUCP (j eric townsend) says:
> Isn't this in essence what CBM did with the "sprite"s on the Commodore
> 64?  (Can I mention that machine in this group?  I didn't think so. :-)
> ...
> Anybody know what home/micro computer did it first?  I'm pretty
> sure the Vic-20 could do this, but I'm not sure about the PET.
> -- 
> Skate UNIX or go home, boogie boy...
>      [Obscure joke goes here]
> J. Eric Townsend ->uunet!nuchat!flatline!erict smail:511Parker#2,Hstn,Tx,77007
>              ..!bellcore!tness1!/

	The first home system I know of with "sprites" was the Atari
2600 game machine.  They called them "players".  They were very limited
by the hardware; 128 bytes of RAM, 4K ROM, no DMA so the CPU had to do
everything in real time.  The first computers that had this capability
were the Atari 400 and 800, the machines CBM targeted with the C64.
They had 4 "players" and 4 "missles" (2 pixel wide sprites), and all of
them were DMA driven.  The chipset used in the Atari 400 and 800 was
designed by Jay Minor, the same person who designed the chipset used 
in the Amiga.

 ------------------------------------------------------------------------------
			Dan Moore
			AT&T Denver
			ihnp4!druhi!dlm
 ------------------------------------------------------------------------------